Recurrent pregnancy loss is associated with the presence of autoantibodies against phospholipid (PL) antigens in some patients. Mechanisms underlying the development of such autoimmune pregnancy loss are not well understood. Recent reports have suggested existence of T cells that recognize PL antigens bound to an antigen presenting molecule, CD1d. My hypothesis is that such CD1d-restricted PL-reactive T cells induce the production of anti-PL autoantibodies and pregnancy loss associated with these antibodies. In support of this hypothesis, we have found that CD1d-knockout autoimmune-prone (NZB X NZW) F1 mice have significantly reduced serum levels of anti-PL antibodies as compared to their wild-type littermates, suggesting a possible role of CD1d in the development of anti-PL antibodies. Building on my expertise in the biology of CD1d-restricted glycolipid-reactive T cells and based on the above observations, I will test this hypothesis in three Specific Aims. In Aim 1, I'll test the hypotheses that mice with spontaneous or induced anti-PL antibodies have increased numbers and/or activation of PL-reactive T cells in lymphoid organs; such T cells will infiltrate the pregnant uterine mucosa, called decidua, in increased numbers. Then, I'll determine the effect of CD1d-restricted PL-reactive T cells on the production of anti-PL antibodies by B cells in vitro in Aim 2. In Aim 3, I'll determine the effect of CD1d-restricted PL-reactive T cells on pregnancy outcome. We will further investigate whether PL-reactive T cells directly induce pregnancy loss or whether pregnancy loss is mediated via anti-PL antibodies induced by PL-reactive T cells. It is hoped that this study will elucidate a novel pathogenetic mechanism of recurrent pregnancy loss in autoimmune diseases. The data obtained will also form the basis for my first R01 or another extramural proposal.